NMTCB(CT) cheat sheet

Screen every patient before contrast administration for prior reaction, asthma, renal disease, diabetes with metformin use, pregnancy, thyroid disease, and heart failure; document the responses.

IV iodinated contrast is generally withheld when eGFR is below 30 mL/min/1.73 m2; eGFR 30-44 proceeds with caution using reduced volume and pre-hydration.

Withhold metformin 48 hours after iodinated contrast when eGFR is under 30 mL/min/1.73 m2 or acute kidney injury is present; restart only after renal function is confirmed stable.

Premedicate prior mild reactors with prednisone 50 mg PO at 13, 7, and 1 hour before contrast plus diphenhydramine 50 mg PO 1 hour before; methylprednisolone 32 mg PO at 12 and 2 hours is an alternative.

Anaphylactoid response: urticaria alone gets an antihistamine; bronchospasm gets inhaled bronchodilator plus IM epinephrine 0.3-0.5 mg of 1:1000; hypotension gets leg elevation, rapid fluids, and epinephrine.

Extravasation: stop the injection, aspirate residual contrast through the cannula, remove the cannula, elevate the limb, apply cold compresses, and document; surgical consultation for volumes over 100 mL, blistering, ulceration, or neurovascular compromise.

Power injection requires a 20-gauge or larger catheter in a large proximal vein with patency confirmed by saline flush; respect the rated flow limit of the catheter.

Oral contrast for abdomen: 500-1000 mL of dilute (2%) barium or iodinated solution 60-90 minutes before scanning; water is used as neutral contrast when positive opacification is unnecessary.

Patient preparation includes removing metal, verifying the order and identity, confirming NPO status when sedation is planned, and assessing ability to lie supine and hold breath.

DLP (mGy·cm) = CTDIvol (mGy) x scan length (cm); effective dose (mSv) = DLP x k, with approximate adult k values of 0.0021 (head), 0.014 (chest), and 0.015 (abdomen).

CTDIvol reports dose per rotation in a standard phantom; the 32 cm phantom approximates the adult body and the 16 cm phantom is used for head and pediatric measurements.

Dose scales linearly with mAs and rotation time, approximately with the square of kVp, and inversely with pitch (dose proportional to 1/pitch).

Optimize with automatic exposure control (z-axis and angular mA modulation), reduced kVp for small-to-medium adults and contrast-enhanced studies, and scan lengths limited to the anatomy of interest.

Noise is inversely proportional to the square root of mAs; iterative reconstruction permits 20-40% mAs or dose reduction without a rise in image noise.

A 0.5 mm lead-equivalent apron attenuates more than 90% of scattered diagnostic x-rays; bismuth breast shields lower surface dose but can increase tube output if AEC compensates and may degrade images.

Scatter exposure falls with the square of distance from the gantry, so step back or leave the room during helical acquisition.

Fetal doses below 50 mGy carry no measurable increase in deterministic effects; consult radiology or medical physics for any pregnant patient and document the dose estimate.

Pediatric optimization follows Image Gently principles: weight-based protocols, lower kVp, reduced mAs, limited number of phases, and avoidance of unnecessary coverage.

Brain: angle the gantry parallel to the orbitomeatal line, 5 mm slices from posterior fossa to vertex; noncontrast acquisition first for acute stroke and trauma.

Chest: arms up, inspiratory breath-hold, 1-5 mm slices by indication; lung window approximately W 1500 / L -600 and mediastinal window approximately W 350 / L 50.

Abdomen: supine, feet first, portal venous phase at 70-90 seconds; range typically from lung bases to symphysis with oral contrast as ordered.

CTA timing uses bolus tracking with an ROI in the aorta or pulmonary trunk and a trigger threshold of 100-150 HU, followed by a 30-50 mL saline chaser at the same flow rate.

Positioning landmarks: xiphoid approximates T9, iliac crest approximates L4, and the patient is centered at gantry isocenter for accurate AEC performance.

Liver segments I-VIII (Couinaud) are defined by the hepatic and portal veins; segment I is the caudate lobe, lying between the IVC and the ligamentum venosum.

Renal phases: corticomedullary 30-40 seconds, nephrographic 90-120 seconds, and excretory 5-10 minutes; the right kidney sits lower than the left because of the liver.

HRCT chest uses 1-1.25 mm collimation with a high-spatial-frequency kernel plus inspiratory, expiratory, and prone images to demonstrate air trapping and dependent atelectasis.

The circle of Willis is an arterial ring at the brain base formed by the anterior, middle, and posterior cerebral arteries with anterior and posterior communicating arteries.

HU = 1000 x (mu tissue - mu water) / (mu water - mu air); water is 0 HU, air is -1000 HU, and compact bone is roughly +1000 HU or higher.

Window width sets the displayed HU range and therefore contrast; window level sets the center HU and therefore brightness (lung W ~1500, bone W ~2000, brain W ~80).

Noise decreases with the square root of mAs; spatial resolution improves with a smaller focal spot, thinner slices, and sharper reconstruction kernels.

Helical pitch = table travel per gantry rotation divided by total collimated beam width; detector rows times collimation gives coverage per rotation (64 x 0.625 mm = 40 mm).

Common artifacts: beam hardening (cupping, dark streaks), photon starvation (shoulders), metal streak, motion, partial volume, and ring artifact from detector calibration failure.

Post-processing from thin isotropic data includes multiplanar reconstruction, maximum and minimum intensity projection, and volume rendering.

Daily QA includes tube warm-up and air calibration; water phantom CT number must stay within about plus or minus 5 HU, with noise and uniformity also checked.

Annual physics testing uses the ACR phantom to evaluate CT number accuracy, uniformity, low-contrast detectability, spatial resolution, and slice thickness.

Contrast resolution depends on noise, slice thickness, and dose, whereas spatial resolution depends on focal spot size, detector aperture, and reconstruction kernel.